New polymers with superior performance, unique properties, improved cost effectiveness, quality, and low ecological impact are of ongoing interest to industry. In particular, functional polymers having branched, compact structures and terminally located reactive groups are in demand because of their lower inherent viscosity and higher reactivity as compared to conventional linear statistical co-polymers.
Such superior polymers can be prepared by copolymerizing hyperbranching hydroxycarboxylic acid comonomers (hyperbranching ABn type, where A and B are moieties with hydroxyl- or carboxyl-derived reactive groups, n is 2 or more) (Hult et al., pp. 656-658 and Voit et al., pp. 658-659 in Concise Polymeric Materials Encyclopedia, ed. J. C. Salomone, CRC Press, New York, 1999) and a variety of linear hydroxycarboxylic acid comonomers (linear AB type), including 3-hydroxyvaleric acid (3-HVA) and 3-hydroxybutyric acid (3-HBA).
3-Hydroxycarboxylic acid is useful as a (co)monomer for making linear polyesters. Polyesters are useful as thermoplastic, thermoset, semicrystalline, amorphous, rigid, and elastomeric materials. They are the basis of fibers, films, moldings, and coatings (Goodman, pp. 793-799 in Concise Encyclopedia of Polymer Science and Engineering, ed. J. I. Kroschwitz, John Wiley & Sons, New York, 1990).
3-Hydroxyvaleric acid has been prepared by the 3-hydroxylation of valeric acid in fermentation using Candida rugosa (Hasegawa et al., J. Ferment. Technol. 59:257-262 (1981); JP 59053838 B4), and a single enantiomer of 3-hydroxyvaleric acid was similarly prepared by fermentative 3-hydroxylation of valeric acid with Pseudomonas putida, Pseudomonas fluorescens, Arthrobacter oxydans and Arthrobacter crystallopietes (U.S. Pat. No. 3,553,081). These methods for fermentative oxidation of valeric acid typically produce 3-hydroxyvaleric acid at low product concentrations, and require an elaborate and expensive separation of 3-hydroxyvaleric acid from the fermentation broth. (R)-(−)-3-Hydroxyvaleric acid has been prepared by the chemical degradation (Seebach et al., Helv. Chim. Acta 77:2007-2034 (1994)) or by fermentative autodegradation (WO 9929889) of poly(3-hydroxybutyrate/3-hydroxyvalerate), but degradation of hydroxybutyric acid/hydroxyvaleric acid copolymers also requires a difficult separation of 3-hydroxybutyric acid from the co-product 3-hydroxyvaleric acid. (R)-(−)-3-Hydroxyvaleric acid has also been prepared by the enzymatic reduction of 3-oxovaleric acid (Bayer et al., Appl. Microbiol. Biotechnol. 42:543-547 (1994)) or by the asymmetric hydrogenation of methyl 3-oxovalerate followed by saponification (Burk et al., Organometallics 9:2653-2655 (1990)).
Nitriles are readily converted to the corresponding carboxylic acids by a variety of chemical processes. These processes typically require strongly acidic or basic reaction conditions and high reaction temperatures, and usually produce unwanted byproducts and/or large amounts of inorganic salts as unwanted waste. Reaction conditions for the chemical hydrolysis of nitrites that additionally have a hydroxyl group, such as for the conversion of 3-hydroxyvaleronitrile to 3-hydroxyvaleric acid, will usually result in the undesirable elimination of primary, secondary, or tertiary hydroxyl groups to produce carbon-carbon double bonds.
The enzyme-catalyzed hydrolysis of nitrites to the corresponding carboxylic acids is often preferred to chemical methods, since the reactions are often run at ambient temperature, do not require strongly acidic or basic reaction conditions, and produce the desired product with high selectivity at high conversion. A combination of two enzymes, nitrile hydratase and amidase, can be used to convert aliphatic nitrites to the corresponding carboxylic acid in aqueous solution. The aliphatic nitrile is initially converted to an amide by the nitrile hydratase, then the amide is subsequently converted by the amidase to the corresponding carboxylic acid. A wide variety of bacterial genera are known to possess a diverse spectrum of nitrile hydratase and amidase activities (Sugai et al., Biosci. Biotech. Biochem. 61:1419-1427 (1997)), including Rhodococcus, Pseudomonas, Alcaligenes, Arthrobacter, Bacillus, Bacteridium, Brevibacterium, Corynebacterium and Micrococcus. The fungus Fusarium merismoides TG-1 has also been used as catalyst for the hydrolysis of aliphatic nitrites and dinitriles (Asano et al., Agric. Biol. Chem. 44:2497-2498 (1980)). Immobilized nitrile hydratase and amidase from Rhodococcus sp. (SP409 from Novo Industri) was used to hydrolyze 3-hydroxypropionitrile, 3-hydroxyheptanenitrile, and 3-hydroxynonanenitrile to the corresponding 3-hydroxycarboxylic acids in 63%, 62% and 83% yields, respectively (de Raadt et al., J. Chem. Soc. Perkin Trans. 1, 137-140 (1992)). The formation of the corresponding amide was also observed by TLC. In contrast, the purified nitrile hydratase of Bacillus pallidus Dac521 hydrolyzed a variety of aliphatic nitriles, but did not hydrolyze 3-hydroxypropionitrile (Cramp et al., Biochim. Biophys. Acta 1431:249-260 (1999)).
A single enzyme, nitrilase, also converts a nitrile to the corresponding carboxylic acid and ammonia in aqueous solution, but without the intermediate formation of an amide (Equations 1 and 2).

Kobayashi et al. (Tetrahedron 46:5587-5590 (1990) and J. Bacteriology 172:4807-4815 (1990)) have described an aliphatic nitrilase isolated from Rhodococcus rhodochrous K22 which catalyzed the hydrolysis of a variety of aliphatic nitriles to the corresponding carboxylic acids. A nitrilase from Comamonas testosteroni has been isolated that can convert a range of aliphatic α,ω-dinitriles to either the corresponding ω-cyanocarboxylic acids or dicarboxylic acids (CA 2,103,616; Lévy-Schil et al., Gene 161:15-20 (1995)). Aliphatic nitrilases are also produced by Rhodococcus rhodochrous NCIMB 11216 (Bengis-Garber et al., Appl. Microbiol. Biotechnol. 32:11-16 (1989); Gradley et al., Biotechnology Lett. 16:41-46 (1994)), Rhodococcus rhodochrous PA-34 (Bhalla et al., Appl. Microbiol. Biotechnol. 37:184-190 (1992)), Fusarium oxysporum f. sp. melonis (Goldlust et al., Biotechnol. Appl. Biochem. 11:581-601 (1989)), Acinetobacter sp. AK 226 (Yamamoto et al., Agric. Biol. Chem. 55:1459-1473 (1991)), Alcaligenes faecalis ATCC 8750 (Yamamoto et al., J. Ferment. Bioeng. 73:425-430 (1992)), and Acidovorax facilis 72W (Gavagan et al., J. Org. Chem. 63:4792-4801 (1998)).
The gene encoding the A. facilis 72W (ATCC 55746) nitrilase has been cloned and recombinantly expressed (WO 01/75077) and Chauhan et al., Appl Microbiol Biotechnol, 61:118-122 (2003)). The A. facilis 72W nitrilase converts 3-hydroxynitriles to the corresponding 3-hydroxycarboxylic acids in high yield (U.S. Pat. No. 6,562,603). The 3-hydroxyacids produced are useful as substrates for the preparation of polymers including highly branched copolyesters. Two particularly useful 3-hydroxycarboxylic acids are 3-hydroxyvaleric acid (3-HVA) and 3-hydroxybutyric acid (3-HBA). Nitrilases having improved nitrilase activity for the conversion of 3-hydroxyvaleronitrile (3-HVN) or 3-hydroxybutyronitrile (3-HBN) to the corresponding 3-hydroxycarboxylic acid in high yield at up to 100% conversion would be very useful for reducing industrial production costs.
The problem to be solved, therefore, is to provide nitrilase enzymes with activity useful for converting 3-hydroxynitriles to their corresponding carboxylic acids in high yield. More specifically, nitrilase enzymes possessing a significant improvement in nitrilase activity (relative to the nitrilase activity of the A. facilis 72W) for the conversion of 3-hydroxynitriles (e.g., 3-hydroxyvaleronitrile or 3-hydroxybutyronitrile) to the respective 3-hydroxycarboxylic acids would be useful for reducing industrial production costs.